运用证据权重与确定性系数合成模型进行钉螺-环境因子的空间关系研究

钉螺是血吸虫唯一中间宿主,研究钉螺与孳生环境因子的空间关系对采取有效的查螺、灭螺措施有重要的科学和社会价值。以鄱阳湖恒湖农场茶叶港草洲为研究对象,应用证据权重法与确定性系数法合成模型研究植物南荻、苔草、藜蒿、土壤温度、土壤湿度、导电率、海拔高程7个证据因子与"钉螺出现与否"的空间关系。结果表明:研究区内一定范围内高程的增加易于钉螺孳生;钉螺在湿度相对较大(大于0.35%(m~3/m~3))的环境中分布较少;苔草丰度过大(高于4 368 cm~3/m2)过小(低于1 457 cm~3/m2)都不利于钉螺孳生;钉螺较易孳生在南荻丰度较大(高于2 745 cm~3/m2)的区域;电导率较高(高于0.07 m S/cm)区域钉螺较少;土壤温度在16~20℃范围利于钉螺孳生。在模型预测结果图中,占研究区总面积13.7%的极高易发区的较小范围内预测到82%的钉螺空间分布,评价曲线的曲线下面积高达88.7%,说明模型的预测成功率很高,表明该模型适合研究区域钉螺孳生与其影响因子之间的空间关系。

APPLICATION OF SYNTHETIC MODEL OF WEIGHTS OF EVIDENCE AND CERTAINTY FACTOR TO STUDY OF SNAIL-ENVIRONMENTAL FACTOR SPATIAL RELATIONS

Oncomelania snails are important intermediate hosts of schistosome.The study of "snail-environment" relations has important scientific and social value for the snails' investigation and elimination.This study was a case research performed at Chayegang marshland near Henghu farm (Xinjian County) in the Poyang Lake basin.An incorporating model of weights of evidence and certainty factor was used to explore if the snail's occurrences were affected by seven factors,which including three types of plant (Carex cinerascens,Miscanthus sacchariflorus,and Artemisia selengensis) coverage,soil temperature,soil humidity,soil electrical conductivity,sampling sites elevation,etc.The results showed the increasing site elevation might benefit the snails producing,and snails were very few if the soil humidity was larger than 0.35％ (m3/m3),or the Carex cinerascens productions too large (＞4 368 cm3/m2) or too small (＜ 1 457 cm3/m2),or the soil electrical conductivity larger than 0.07 ms/cm.However,the snails preferred the circumstances if the Miscanthus sacchariflorus productions were larger than 2 745 cm3/m2,or the soil temperature varied between 16℃ and 20℃.The model projection results showed that 87％ snails mostly distributed in the 13％ studied area that snails preferred.The model projected Area Under Curve (AUC) could cover 88.7％ of the snail's real distribution areas,which indicated the model's high prediction abilities,and the model can be applied to the studies of regional Oncomelania snails distribution and the environmental factors.